search

ryanjulian / rllab

rllab is a framework for developing and evaluating reinforcement learning algorithms, fully compatible with OpenAI Gym.
Other
1 stars 21 forks source link

Implement specific reward function, goal sample function, get observa… #106

Closed hejia-zhang closed 6 years ago

hejia-zhang commented 6 years ago
  1. make every specific task env has its own implementation of sample_goal, get_observation, reward, _goal_distance.